Iris Murdoch claims that pursuing goodness is “pointless.” She means that the attempt to act rightly, in accordance with virtue, say, cannot be given any external justification. In other words, such actions have no goal beyond themselves, however helpful they may happen to be. This echoes the Aristotelian dictum that virtuous action is undertaken for its own sake. In this case, an internal justification, which appeals to other ethical notions, would be available. So, we might recommend acting courageously because that would be the wise thing to do. But both courage and wisdom are already normative concepts: they are already replete with ethical normative authority. In this way, no attempt is made to justify specific ethical claims through appeal to non-ethical foundations. Ethical justification is presented as a closed circle.
One response to this kind of circularity would be to charge the Murdochian agent with dogmatism. This naturally leads to another suspicion, namely the sceptical doubt that if the only justifications for acting ethically are themselves already ethicised, then perhaps no genuine ethical justification is to be had. But such responses would be misguided; they reflect an unwarranted demand for foundations, for an Archimedean point outside of the activities of justifying, reasoning and communicating with one another, from where we can issue guarantees for them. Whether we are realists or not (and both myself and Murdoch are), demanding such guarantees is not cautious but pathological; in some lights, it verges on the autistic.
One of the problems met in attempting to give an external justification for genuinely following norms (rather than merely helpful conventions) is that it invites us to answer ill-formed questions. There is something incoherent about questions like ‘what reason do I have to be rational?’ or ‘should I do what I ought to do?’ when they are directed at normativity in general rather than the justification of specific norms. For any answer to these questions to move us, we must already be trading in reasons, which threatens to make any answer seem either hopelessly circular or entirely redundant.
We find a deep affinity here between these awkward questions, asking about a norm for following norms, or a reason to be rational, and the so-called problem of the ‘Kantian paradox’. If we create or legislate normative standards for our actions, there is a difficulty in finding norms for this legislation itself which would prevent this legislation from being enirely sporadic and arbitrary. In other words, we would already need norms to guide the institution of norms. Similar problems loom here to those above, since if there were already authoritative norms to appeal to then self-legislation will be redundant, but if there are not then no non-arbitrary legislation can be undertaken. The upshot, I think, is that self-legislation is an incoherent way to think about the ultimate source of normativity.
The Kantian paradox is significant, but unlike Kantian constructivisits, I think that the attempt to provide a straight solution to it is misguided. Instead, it provides us with an important clue to a structural feature of normativity, best accounted for by Murdoch’s considerations about the pointlessness of pursuing goodness. The lesson it teaches us is that there cannot be any justificatory grounding to normative authority which is not itself equally normative and equally groundless: it is normativity ‘all the way down.’ The lack of non-holistic support for normativity does not undermine the importance of normativity; in fact, it is quite the contrary. Pihlström makes this point well in relation to morality:
Morality does not have any external goal or legitimation. Yet, this, instead of sacrificing the moral seriousness emphasized by the moral realist, is an affirmation of such seriousness. Morality is something serious—indeed, the most serious and most important thing in our life, ‘overriding,’ as one often says—precisely because it does not have any external, non-ethical goal or point.
Bearing this claim in mind, one pertinent criticism of constructivism is that it offers the wrong kinds of reason to be moral (to re-purpose Bernard Williams’ expression). For example, when Korsgaard tells us that we should be moral because otherwise we will lose something more valuable than our lives, namely our identities as agents, morality is being anchored to some external goal. But this is to instrumentalise morality, to make it into a hypothetical imperative: if you want to protect an identity precious to you, then follow these instructions. In so doing, we lose our appreciation of the inherent worth of moral action, which sits alongside its subsidiary benefits to our lives, but is not entirely parasicitic upon them. One need not have as rigoristic a conception of ethical life as Kant to think that right action can be inherently worthy. Indeed, this style of criticism has been echoed by Bradley and Prichard (neither of whom were Kantians), who also think there is something wrongheaded about giving reasons to be moral in general.
Moral or ethical normativity is a distinctive variety, insofar as it often connected to the notions of the categorical and the obligatory. However, I think we can extend the same points about there being no external goal or point to being responsive to the force of the better reason, generalising them to normativity as a whole. Naturally, there can be many subsidiary benefits for agents who follow norms, whether this be the fruits of theoretical or practical reason, such as working out what to what will satisfy us and how to get it, for example. But Murdoch is right to emphasise that the true pursuit of the good is “austere and unconsoled.” What distinguishes responsiveness to normativity proper, rather than following useful conventions, is this discerning disregard for any immediate further goal. Normativity does not, in Murdoch’s vocabulary, come with the consolations of purposiveness beyond itself.
Grundlegung 
I totally don’t get what you mean. If morality doesn’t refer to anything outside of itself, then what is it? “I do what I do because I think I should do it”? Help me out here.
In the post, I argue against certain ways of justifying morality as a whole. Some people used to argue that we ought to follow moral norms because they promote our own self-interest or welfare, and promoting our own self-interest is unproblematically reason-giving. Others, like some constructivists, think we ought to follow moral norms because it is the only way to be a genuine agent, or that it is the only way to preserve our moral identities or our autonomy. Again, the idea is that morality can be given some non-moral justification. I think this misses an important feature of moral justification, which Aristotle was getting at when he claimed that the virtuous action was done for its own sake, and Kant was getting at when he claimed that moral action was done solely out of respect for the moral law.
The threat is that all this makes things like moral motivation appear mysterious, or even morality itself, as you seem to be worried. However, I do not think it does. Firstly, we can still tell an explanatory, rather than justificatory, story about the development of moral practices. Secondly, we can still offer what I’ve called internal justifications for specific moral norms (e.g. ‘homophobia is a fundamentally unjust attitude, so do not discriminate on the basis of sexuality’). But it can still seem something is missing here.
What then is morality? My answer here is deflationary. Morality is striving to avoid cruelty, it’s knowing when courage is required of us, it’s promoting justice — all the banal and profound ways of acting in and looking at the world that are already reasonably familiar to us. Doubtless, this sort of answer can seem disappointing. But I don’t think we have any other sort of answer available to us. If you want to know the sort of thing that makes moral action worthwhile, just look to the kinds of things we appeal to in everyday moral discourse. Of course, we can and should argue about what is the right thing to do (and I think there is a correct answer to many such arguments), but there is no philosophical shortcut over and above doing our best to get reality right through exploring the issues in discussion, the arts, imaginative reflection, and so on — just as for the other non-moral areas of our lives.
Good post Tom. However, I think you’re perhaps overstating your case against constructivists. One can agree that the attempt to provide some justification for doing what one ought to do (other than reasons why it is what one ought to do) is wrong-headed (and thus agree with your criticism of Korsgaard) and yet still accept the Kantian autonomy thesis in some form. I’m not sure if what I’m about to say is all that clear (my head is well out of issues about normativity at the moment), but I’ll give it a go anyway.
You’ve argued before that the autonomy thesis leads to the paradox you pose above, namely, if all norms gain force only through some act of binding on the part of the subject, how can such acts of binding be norm-governed? This is indeed a serious problem, but I’m not sure it is commensurate with the problem of the unjustifiability of justification.
The point is really this, although there is some sense in which the autonomy thesis gives us an idea of what kind of reasons there might be for why one ought or ought not to act in specific cases, in terms of what one has said and done commit one to, it needn’t be applied to find the same kind of reasons in the case of the fundamental norms, i.e., those norms that govern the very process of giving and asking for reasons for thought and action, and undertaking commitments generally. These norms can be transcendentally guaranteed. This is what I’ve talked about before in terms of the ‘primary bind’.
At this point you might want to claim that such a transcendental guarantee is precisely the kind of external justification of normativity that you’re arguing against, but I think that this isn’t the case. The transcendental argument does not have the form of ‘insofar as you want to x, you ought to y’, or ‘insofar as you have said/done x, you have committed yourself to y’, but rather that ‘insofar as you do or say anything at all you are committed to these norms’. I think this has the consequence of actually implying the main point you’re making, namely, that one can’t justify justification. Insofar as one is even in a position to raise the question of whether one should act in ways one has reason to act, one is already bound by the force of reasons.
If the fundamental norms of rationality, which govern the process of binding oneself to norms, are transcendental, then they can be taken as a limit-case of the autonomy thesis rather than as posing a paradox (I’m not going to rehearse this point in its entirety, though I know it requires a bit more explanation). The pertinent question is why would we adopt this transcendentally inflected version of the autonomy thesis if we can simply claim outright that there is no justification of justification, that one simply ought to do what one ought to do, for no further reason. The reason is that although the autonomy thesis doesn’t help us ground normativity in something non-normative, it does help us understand the structure of normativity itself. However, it doesn’t allow us to reduce normativity to non-normative terms. We must give an account of how the structure of the normative are grounded in practices that are themselves norm-governed, specifically, governed by fundamental norms.
Thanks very much for your comprehensive and patient explanation – I can see that my initial read of your post was quite sloppy.
I’m definitely not trying to be contentious, but don’t you still need something like a “because of” to make what you’ve said in your explanation cohere? What I mean is, isn’t there an implicit “because it feels good” or “because I intuit a certain ineffable rightness in so acting” behind or alongside the deflated, “non-justificatory” non-reasons you cite for acting morally?
I appreciate your time, thank you.
Thanks for your comment Ricky.
There might be two worries lurking here. Firstly, it might be a concern with moral motivation, and so what could get people to act morally — must it be that they get some sort of pleasure out of doing so; could it be that they can decide to follow some sort of ethical intuition? Secondly, it might be a concern with the justification of moral actions — what makes them moral (or at least the sort of thing we ought to do)? Perhaps the answer to the second question might have something to do with them being pleasureable or promoting pleasure, or again, maybe it is a kind of ineffable fact (which we may then be able to somehow intuit).
Both issues raise the sort of ‘because of’ troubles you gesture towards. If we take the motivation case first, I think that we can be motivated by a recognition that the situation we find ourselves in demands a certain action from us. But I don’t think anything too spooky goes on here, as if we had some special faculty for divining moral facts. Instead, I think our ordinary experience is rich enough to support an ethical dimension to the way the world appears to us: we feel that the jab in ribs is cruel, say, or we immediately recognise that standing up to the attacker is noble, just as we do with other experiences with an anthropocentric dimension, such as colour experiences.
Often this sort of view is treated with suspicion. Alan Thomas (who has a broadly similar view to me) has remarked that
I think such suspicion is unfair, especially when the ethical experience at hand is conflated with a kind of moral sense which allows direct access to non-natural moral truths, which I agree is a rather crazy thing to believe.
Anyway, I think that there is an understanding of the ethical dimension of experience whereby, alongside the right kind of concepts and reflective capacities, it can reveal the world to us in a way that allows us to be motivated by what we experience. Obviously, there’s lots more that could be said here, and basically I am in broad agreement John McDowell on these issues (especially his seminal paper ‘Virtue and Reason’).
The second issue, namely justification, is probably more pressing though. Without some sort of understanding of what makes an action moral in the first place, it can seem that we’ve no idea what we’d even be experiencing in ethical experience. This issue is rather thorny, especially as I subscribe to a kind of non-standard realism about normativity, and have a Wittgensteinian suspicion about giving a general theory of morality (or normativity itself).
In massively compressed form then, I think that in some sense the situations we find ourselves in themselves determine what we are morally required to do. In a sense, the answer to the ‘because of’ question is ‘because of the world.’ So, moral actions are those which our rational responsiveness to the world requires us to do, and which are usually categorically binding and override other reasons which we have.
However, for various reasons, things are massively more complicated than that on my view. For one, saying ‘because of the world’ is not a comprehensive answer, because I don’t think we can specify what it is about the world that makes it normatively compelling without presupposing our existing moral practices and evaluative judgements. As I say in the post, ethical justification is a kind of closed circle. We need to retain an anthropocentric perspective, including our existing evaluative perspective, when we are characterising what morality does and does not require. Whilst our perspectives will be anthropocentric in this way, this can be so without impugning the reality of what they reveal though. So, it gets kind of complex. I’ve gone on long enough for now, but hopefully that’ll clarify things a little bit, albeit with lots more potentially puzzling bits along the way too.
Pete:
I didn’t mean to indict all versions of constructivism here, only those that try to directly ground morality in something like the conditions of agency (or an autonomy which arises from these conditions), and those that fall into ‘reasons for rationality’ problems. I agree that not all versions of constructivism do this, nor all uses of the autonomy thesis. I didn’t have your kind of transcendental vindication of fundamental rational norms in mind, for example. So too, perhaps my comparisons between the Kantian paradox and the ‘reasons for rationality’ predicament are a bit overdrawn. Both seem to problematically require norms for the vindication of norms, from which I draw a realist lesson, but I agree that answering the Kantian paradox need not embroil someone in an attempt to give reasons to be rational.
As an aside, it might be worth mentioning that Brandom seems to address a problem in this area — whether we have a reason to self-legislate at all — in his second Woodbridge lecture:
I’m drawn in to your blog via an old post on Brandom, but I like what I’m seeing so much I think I’ll stick around and see what else you have to say.
I wonder what your thoughts would be on a minimalism about normativity, as opposed to a pluralism. Specifically, I am curious as to whether you think all the virtues–or all the sources of value–must be justified ultimately by this basic notion you are talking about in this post that comes about just from our being able to give and ask for reasons, trying to understand each other, and attempting to communicate and cooperate or whether you think there might be a case for pluralism, that there might be distinct sources of virtue or value that are not ultimately justified by or reduced to that basic notion. I’m inclined to think with Socrates that pluralism about value is false (or, at least, not unavoidable), but I’d like to know what you think of this suggestion.
Also, thanks for mentioning that McDowell paper. I can’t believe I have never been told of this one before!
Just to clarify — as my language has been a little sloppy in the above — that it is the activity of assessing the authority of norms which I think must always draw on anthropocentric concepts and cannot alight upon a non-circular foundation. So, I don’t think that playing what Brandom calls ‘the game of giving and asking for reasons’ is itself the primary source of normative authority. I try to hold onto the realist idea that reasons are there anyway, products of situations themselves, rather than being products of our recognition of them. But at the same time I try to deflect any platonic reading of that claim, attempting to domesticate it so that it’s acceptable in a broadly naturalistic world view, albeit not a narrowly scientistic one. This is the same kind of strategy McDowell pursues too.
I’m not sure what I think about value pluralism. I believe that there can be tragic conflicts, where there are situations which involve an unresolvable clash of genuinely valuable things, and this suggests some affinity with pluralism. (I often think appeals to ‘reasonable pluralism’ in the context of liberal political philosophy are amiss though.) When it comes to rationality, I’m often tempted by the idea that practical reason is a kind of unitary ability, not a distinct set of capacities to recognise reasons (although I don’t really buy McDowell’s argument for the unity of the virtues). I guess some kind of pluralism about value and monism about our responsiveness to it can be made compatible though.
I love that McDowell article, by the way: it’s definitely a contender for my favourite philosophy paper!